Hex and int Kconfig options are supposed to have defaults. This is so we
can configure U-Boot without having to enter particular values for the
items that don't have specific values in the board's defconfig file.
If this rule is not followed, then introducing a new Kconfig can produce
a loop like this:
Break things (BREAK_ME) [] (NEW)
Error in reading or end of file.
Break things (BREAK_ME) [] (NEW)
Error in reading or end of file.
The continues forever since buildman passes /dev/null to 'conf', and
the build system just tries again. Eventually there is so much output that
buildman runs out of memory.
We can detect this situation by looking for a symbol (like 'BREAK_ME')
which has no default (the '[]' above) and is marked as new. If this
appears multiple times in the output, we know something is wrong.
Add a filter function for the output which detects this situation. Allow
it to return True to terminate the process. Implement this termination in
cros_subprocess.
With this we get a nice message:
buildman --board sandbox -T0
Building current source for 1 boards (0 threads, 32 jobs per thread)
sandbox: w+ sandbox
+.config:66:warning: symbol value '' invalid for BREAK_ME
+
+Error in reading or end of file.
+make[3]: *** [scripts/kconfig/Makefile:75: syncconfig] Terminated
+make[2]: *** [Makefile:569: syncconfig] Terminated
+make: *** [Makefile:177: sub-make] Terminated
+(** did you define an int/hex Kconfig with no default? **)
Signed-off-by: Simon Glass <sjg@chromium.org>
If the process outputs a lot of data on stdout this can be quite slow,
since the bytestring is regenerated each time. Use a bytearray instead.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present cros_subprocess and the tools library use a string to obtain
stdout from a program. This works fine on Python 2. With Python 3 we end
up with unicode errors in some cases. Fix this by providing a binary mode,
which returns the data as bytes() instead of a string.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present this function uses lists and strings. This does not work so
well with Python 3, and testing against '' does not work for a bytearray.
Update the code to fix these issues.
Signed-off-by: Simon Glass <sjg@chromium.org>
The communication filter reads data in blocks and converts each block to
unicode (if necessary) one at a time. In the unlikely event that a unicode
character in the input spans a block this will not work. We get an error
like:
UnicodeDecodeError: 'utf8' codec can't decode bytes in position 1022-1023:
unexpected end of data
There is no need to change the input to unicode, so the easiest fix is to
drop this feature.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
os.read() returns a byte array in Python 3.5.2 and needs to be converted
into a string. Check if the returned value is an instance of bytes and
if it is decode it as a utf-8 string. If it is not a utf-8 encoded string
the decoding may fail with an exception.
Prior to this fix the comparisions check data == "" would fail when data
was b'' and would cause an infinite memory leaking loop. joins would
also fail with an exception below but due to the infinite loop it never
made it that far.
Signed-off-by: George McCollister <george.mccollister@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Syntax for exception handling is a little more strict in python 3.x.
Convert all uses to a form accepted by both python 2.x & python 3.x.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Acked-by: Simon Glass <sjg@chromium.org>
This adds a new library on top of subprocess which permits access to
the subprocess output as it is being generated. We can therefore
give the illusion that a process is running independently, but still
monitor its output so that we know what is going on.
It is possible to display output on a terminal as it is generated
(a little like tee). The supplied output function is called with all
stdout/stderr data as it arrives.
Signed-off-by: Simon Glass <sjg@chromium.org>
